AI GitHub
LlamaIndex for connecting documents to model queries
Build ingestion, indexing and query workflows around external data.
Key Takeaways
- LlamaIndex defines document and node abstractions and provides components for parsing, retrieval and response synthesis.
- Choose the components that read sources, create nodes and generate embeddings.
- Index a small documentation set and check the retrieved nodes before reading the answer.
- A model can answer correctly from prior knowledge while the index retrieves the wrong text.
How it works
LlamaIndex defines document and node abstractions and provides components for parsing, retrieval and response synthesis. Individual stages can be replaced as the application matures.
What you need
Choose the components that read sources, create nodes and generate embeddings. If you plan to cite pages, carry source identity into the index before experimenting with answer synthesis.
What to try first
Index a small documentation set and check the retrieved nodes before reading the answer. Change only the chunking configuration and measure how citations move.
Store the original page identifier on every chunk. If an answer is wrong, the identifier helps trace it to an extraction, indexing or generation problem.
Where to look for failures
A model can answer correctly from prior knowledge while the index retrieves the wrong text. Evaluate retrieval and final answers separately.
Related Franklin pages
Test conversion, indexing and retrieval independently before connecting a generator. Keep an original page reference on each result so you can debug a bad answer.
- Wire document stores, retrievers, rankers and generators into explicit pipelines. Haystack for component-based retrieval pipelines
- Extract document structure, reading order and tables for search or analysis. Docling for structured document conversion
For the full path from source documents to generated answers, start with Franklin’s glossary. Retrieval-augmented generation
Project documentation
Try it for your workflow
Explore the project’s documentation for setup and usage, or save this page to a collection for later.
Read the project documentation ↗